/**
* Calculates the distance between this voice sample and the voice prints previously extracted
* and returns the closest matches sorted by distance
* <p>
* Usage of a closed set is assumed : the speaker's voice print was extracted before and is known to the system.
* This means you'll always get MatchResults even if the speaker is absolutely unknown to the system.
* The MatchResult class provides a likelihood ratio in order to help determining the usefulness of the result
* </p>
* @param voiceSample the voice sample, values between -1.0 and 1.0
* @return a list MatchResults sorted by distance
*/
public IEnumerable <MatchResult <T> > Identify(double[] voiceSample)
{
if (store.Count == 0)
{
throw new InvalidOperationException("There is no voice print enrolled in the system yet");
}
var voicePrint = new VoicePrint(audioProcessor.ProcessAndExtract(voiceSample));
var calculator = new EuclideanDistanceCalculator();
var matches = new List <MatchResult <T> >(store.Count);
var distanceFromUniversalModel = voicePrint.GetDistance(calculator, universalModel);
foreach (var entry in store)
{
var distance = entry.Value.GetDistance(calculator, voicePrint);
// likelihood : how close is the given voice sample to the current VoicePrint
// compared to the total distance between the current VoicePrint and the universal model
int likelihood = 100 - (int)(distance / (distance + distanceFromUniversalModel) * 100);
matches.Add(new MatchResult <T>(entry.Key, likelihood, distance));
}
matches = matches.OrderBy(m => m.Distance).ToList();
return(matches);
}
public void TestEuclideanDistanceBetweenElementsHavingIdenticalCoordinates()
{
var calculator = new EuclideanDistanceCalculator();
var element = new ClusterPoint { X = 0, Y = 0 };
var otherElement = new ClusterPoint { X = element.X, Y = element.Y };
var distance = calculator.CalculateDistance(element, otherElement);
Assert.AreEqual(0.0, distance, double.Epsilon, @"The distance must be 0!");
}
public static void Main(string[] args)
{
string filePath = "sunflower.jpg";
string outputFile = "result.jpg";
int classNumber = 5;
IImageBinarizer imageBinarizer = new MinimumErrorThresholder();
IImageProcessor imageProcessor = new ImageProcessor(filePath);
ILabelingService labelingService = new IterativeLabelingService();
IDistanceCalculator<ClassificationUnit> euclidian = new EuclideanDistanceCalculator<ClassificationUnit>();
var classificationService = new ClassificationService(imageBinarizer, imageProcessor, labelingService, euclidian, classNumber);
classificationService.Classify();
classificationService.Save(outputFile);
}
public void Calculate_Calculates_EuclideanDistance()
{
var loc_a = new Location()
{
Longitude = 48.8584, Latitude = 2.2945
};
var loc_b = new Location()
{
Longitude = 40.6413, Latitude = -73.7781
};
var distance = new EuclideanDistanceCalculator().Calculate(loc_a, loc_b);
var expected = 76.5151044119395;
Assert.Equal(expected, distance, 5);
}
public static double CalcDiagonalYZ(double width, double height)
{
double distance = EuclideanDistanceCalculator.CalcDistance2D(0, 0, width, height);
return(distance);
}
